Port HalfFloatRow_SSE2 to AVX2 but not using F16C.

source/planar_functions.cc

 /*
  *  Copyright 2011 The LibYuv Project Authors. All rights reserved.
  *
  *  Use of this source code is governed by a BSD-style license
  *  that can be found in the LICENSE file in the root of the source
  *  tree. An additional intellectual property rights grant can be found
  *  in the file PATENTS. All contributing project authors may
  *  be found in the AUTHORS file in the root of the source tree.
  */
 
@@ skipping 2552 matching lines @@
   }
 #if defined(HAS_HALFFLOATROW_SSE2)
   if (TestCpuFlag(kCpuHasSSE2)) {
     HalfFloatRow = HalfFloatRow_Any_SSE2;
     if (IS_ALIGNED(width, 8)) {
       HalfFloatRow = HalfFloatRow_SSE2;
     }
   }
 #endif
 #if defined(HAS_HALFFLOATROW_AVX2)
-  if (TestCpuFlag(kCpuHasAVX2) && TestCpuFlag(kCpuHasF16C)) {
+  if (TestCpuFlag(kCpuHasAVX2)) {
     HalfFloatRow = HalfFloatRow_Any_AVX2;
     if (IS_ALIGNED(width, 16)) {
       HalfFloatRow = HalfFloatRow_AVX2;
     }
   }
 #endif
+#if defined(HAS_HALFFLOATROW_F16C)
+  if (TestCpuFlag(kCpuHasAVX2) && TestCpuFlag(kCpuHasF16C)) {
+    HalfFloatRow = HalfFloatRow_Any_F16C;
+    if (IS_ALIGNED(width, 16)) {
+      HalfFloatRow = HalfFloatRow_F16C;
+    }
+  }
+#endif
   for (y = 0; y < height; ++y) {
     HalfFloatRow(src_y, dst_y, scale, width);
     src_y += src_stride_y;
     dst_y += dst_stride_y;
   }
   return 0;
 }
 
 // Apply a lumacolortable to each ARGB pixel.
 LIBYUV_API
@@ skipping 379 matching lines @@
     }
     free_aligned_buffer_64(rows);
   }
   return 0;
 }
 
 #ifdef __cplusplus
 }  // extern "C"
 }  // namespace libyuv
 #endif